Alison Frankel

When U.S. Senior District Judge Jed Rakoff rejected a $285 million settlement between Citigroup and the Securities and Exchange Commission last fall, he offered a stern rebuke to SEC lawyers who’d suggested his role was not to protect the public interest. “A court, while giving substantial deference to the views of an administrative body vested with authority over a particular area, must still exercise a modicum of independent judgment in determining whether the requested deployment of its injunctive powers will serve, or disserve, the public interest,” Rakoff wrote in his oft-quoted ruling. “Anything less would not only violate the constitutional doctrine of separation of powers but would undermine the independence that is the indispensible attribute of the federal judiciary.”

In the months since, at least three other federal judges have cited Rakoff in questioning whether settlements proposed by federal agencies serve the public interest, two in SEC cases and one in a Federal Trade Commission case. The SEC adopted a minor, mostly cosmetic revision in the policy that so provoked Rakoff — in which defendants are permitted to settle without admitting liability — but otherwise insisted that such compromises are the very foundation of federal enforcement efforts.

On Thursday, the agency’s position received a very powerful endorsement from the 2nd Circuit Court of Appeals. A three-judge panel ruled that the SEC’s case should be stayed pending a joint appeal of Rakoff’s ruling by the agency and Citigroup, overturning a Rakoff order that the case proceed. The extraordinary 17-page appellate ruling concludes that Citi and the SEC are likely to succeed in their argument that Rakoff was wrong to reject the settlement.

“The S.E.C. asserts that its settlement is in the public interest and that its access to a stay so as to protect the settlement is also in the public interest,” said the per curiam ruling by Judges John Walker, Pierre Leval, and Rosemary Pooler. “We are bound in such matters to give deference to an executive agency’s assessment of the public interest …. We have no reason to doubt the SEC’s representation that the settlement it reached is in the public interest. We see no bases for any contention that the SEC’s decision to enter into the settlement was ‘arbitrary, capricious, an abuse of discretion, or otherwise not in accordance with law.’”

Rakoff “misinterpreted” precedent on his discretion to evaluate the public interest, the appellate judges found, and exceeded his judicial authority: “The responsibilities for assessing the wisdom of such policy choices and resolving the struggle between competing views of the public interest are not judicial ones,” the opinion said.

Over the last six months, U.S. Senior District Judge Jed Rakoff has made Irving Picard of Baker & Hostetler look more like Don Quixote than a white knight riding to the rescue of investors who lost billions in Bernard Madoff’s Ponzi scheme.

Rakoff has already squelched the Madoff trustee’s fraud claims against the banks that allegedly aided and abetted Madoff’s scheme, as well as cutting off Picard clawback claims that date back more than two years. On Tuesday, in Rakoff’s biggest-dollar ruling in the Madoff case, the judge said Picard does not have standing to pursue a $60 billion racketeering suit against UniCredit and two other foreign banks that allegedly participated in a scheme to funnel $9.1 billion to Madoff in exchange for kickbacks to a woman named Sonja Kohn. (Picard had claimed $20 billion in damages, which can be tripled under the Racketeer Influenced and Corrupt Organizations Act.)

Interestingly, Rakoff did not use the same analysis to dismiss the RICO claims as he did in tossing fraud claims against UniCredit and the other banks. In a quick dismissal of the fraud counts, the judge reiterated his July 2011 holding that Picard can’t stand in the shoes of Madoff investors to assert fraud against the financial institutions that allegedly abetted Madoff’s Ponzi scheme. He could have simply said the same is true in the trustee’s racketeering case (which is what I assumed he would do when I wrote about the fraud dismissal in July). Instead, Rakoff seemed to accept, for the purposes of considering the banks’ motion to dismiss, Picard’s argument that Sonja Kohn’s alleged racketeering scheme was, at least to some extent, distinct from Madoff’s scam.

The Solicitor General’s office of the Department of Justice is home to some of the smartest lawyers in the country. These are the people who represent the views of the United States in the most important public policy cases at the U.S. Supreme Court. They go on to head appellate practices at prestigious law firms — or to their own seats in the federal judiciary. Lawyers in the SG’s office are accustomed to deference.

Jed Rakoff, however, isn’t much for deference.

In a Feb. 7 ruling on a claim of attorney-client privilege in a Freedom of Information Act dispute, the Manhattan federal court senior judge conceded that when the Solicitor General’s office makes a representation to a court, “trustworthiness is presumed.” But Rakoff said that when he dug into the SG’s justification for an assertion in a 2009 case at the Supreme Court, he couldn’t find anything aside from some emails exchanged amongst lawyers in the office. “It seems the government’s lawyers were engaged in a bit of a shuffle,” the judge said.

He cited a Peter Finley Dunne aphorism — “Trust everybody but cut the cards” — but might just as well have quoted Ronald Reagan’s famous “Trust, but verify” (itself an adaptation of a Russian proverb favored by Vladimir Lenin). Because Rakoff couldn’t verify the SG’s assertion in the Supreme Court brief, except in emails over which the Justice Department was claiming privilege, he said privilege doesn’t shield parts of the emails.

On Wednesday, the Financial Industry Regulatory Authority disclosed a settlement with Citigroup that U.S. Senior Judge Jed Rakoff might find interesting. Citi agreed to pay a $725,000 fine to resolve allegations that it committed thousands of disclosure lapses in research reports issued between January 2007 and March 2010. (A big thanks to my Thomson Reuters colleague Stuart Gittleman of Accelus, who told me about Citi’s FINRA deal.) Among other disclosure problems, Citi failed to note its role as a manager or co-manager of a related public offering in 8 percent of the 80,000 reports it issued annually; it neglected to report investment banking revenue in 330 research reports; and it didn’t disclose its beneficial ownership in about 1,800 companies its analysts covered.

The FINRA consent letter, signed by Citi counsel Robert Romano of Morgan, Lewis & Bockius, sure makes it sound as though Citi was aware of its disclosure failures. The bank itself identified most of the lapses, which violated strict FINRA disclosure guidelines imposed on Wall Street firms after a 2003 investigation of conflicts of interest in analyst reports. (Citi agreed to pay a $400 million fine after that investigation.) The bank has already conducted two internal reviews of its disclosure systems, one in conjunction with a previous $350,000 fine for lapses committed between 2004 and 2006. In 2010, after continuing problems with internal systems and data from outside affiliates, Citi hired an independent consultant to recommend improvements in its technical disclosure processes. In Wednesday’s consent, the bank agreed (again) to accept a FINRA censure, which is now part of its permanent disciplinary record.

But you won’t find any outright admission of wrongdoing by Citi in Wednesday’s signed consent. To the contrary, the document is sprinkled with the phrase that has become known as Rakoff’s Scourge: “without admitting or denying.” Citi didn’t admit or deny the latest batch of disclosure failures, just as it didn’t admit or deny regulatory allegations in 2003 or alleged disclosure failures in 2006. The latest FINRA consent repeats the boilerplate from Citi’s two previous disclosure agreements with the industry regulator.

Late Friday the Securities and Exchange Commission confirmed in a statement what the New York Times first reported Friday morning: it has changed its policy on the boilerplate “neither admit nor deny” language in most SEC settlement agreements. But don’t get too excited. The change will affect only cases in which the defendant has admitted guilt or been convicted in a related criminal action. In settlements with those criminal defendants, the SEC will delete “inconsistent” concessions and instead “recite the fact and nature of the criminal conviction or criminal [admission] in the settlement documents.”

In other words, defendants whose guilt has already been established under the higher standard of criminal law can no longer evade responsibility for civil charges. Which leads, of course, to the question of why it took the SEC 40 years to change such a ridiculous policy.

In the weeks since U.S. Senior District Judge Jed Rakoff of Manhattan federal court rejected the agency’s proposed $285 million settlement with Citigroup for misleading investors about a synthetic CDO, the SEC has argued long and loud that the boilerplate Rakoff scorned is intrinsic to its ability to reach settlements with defendants worried about liability in follow-on civil suits by private plaintiffs lawyers. I get that. And as I’ve reported, just about every other federal agency with enforcement power has a similar practice of permitting defendants to settle without conceding they’ve done anything wrong. I have my doubts that deleting pro forma “neither admit nor deny” language from settlement agreements would result in a dramatic change in the value of follow-on private settlements, but perhaps I, like most federal judges, have become inured to boilerplate.

Former Goldman Sachs director Rajat Gupta and his lawyer, Gary Naftalis of Kramer Levin Naftalis & Frankel, declared what might seem to be a very strange kind of victory last week when the Securities and Exchange Commission agreed to drop its administrative proceeding against Gupta. The two-page stipulation between Gupta and the SEC makes it clear that the SEC isn’t giving up on its claims that Gupta engaged in insider trading when he allegedly passed confidential information about Goldman Sachs and Procter & Gamble to Galleon Group hedge fund chief Raj Rajaratnam. All Gupta won was a pledge that the agency will sue him in federal court. And that is indeed a huge victory.

The SEC chose an anomalous vehicle for its March 1 suit accusing Gupta of helping Rajaratnam engage in insider trading. Instead of bringing a case against Gupta in Manhattan federal court — as the SEC did when it sued 28 other Galleon insider trading defendants — the agency filed what’s known as a public administrative proceeding. Those proceedings take place before an administrative law judge under SEC rules, not the federal rules of civil procedure. There’s no jury, and the first appeal of any adverse ruling goes to the SEC commissioners, not to an appeals court.

As Gupta counsel Naftalis laid out in his March 18 federal court complaint against the SEC, the differences between an SEC administrative proceeding and a federal court case added up to a big pile of prejudice against Gupta. Gupta wouldn’t be able to take depositions or conduct full discovery to counter the SEC’s assertions. The SEC, meanwhile, would be able to introduce hearsay evidence that it might not be able to get into evidence in federal court, according to the Gupta complaint. “The only plausible inference is that the Commission is proceeding how and where it is against Mr. Gupta for the bad faith purpose of shoring up a meritless case by disarming its adversary,” the complaint asserted.

In the end, it wasn’t even a close call.

Using words like “conjecture,” “bootstrapping,” and “a stretch,” Manhattan federal court judge Jed Rakoff on Thursday decimated trustee Irving Picard‘s multibillion-dollar campaign against the banks that allegedly helped Bernard Madoff engineer his fraud, in a 26-page opinion that left no room for doubt. Rakoff so thoroughly rejected each and every one of Picard’s arguments for why he had the right to bring common law fraud claims against HSBC and UniCredit that the judge didn’t even cite much legal precedent through the first half of the ruling. He simply applied what he calls “ordinary use of the English language” to conclude that no reading of the relevant laws or cases grants Picard standing to sue the banks for unjust enrichment and aiding and abetting fraud and breach of fiduciary duty. This ruling derived its power — and it is a very powerful opinion — from its simplicity.

Rakoff’s ruling immediately affected Picard’s $6.6 billion case against HSBC and a parallel $2.2 billion case against UniCredit. But it’s going to have huge repercussions beyond those suits. Judge Rakoff is also presiding over Picard’s $60 billion racketeering case against UniCredit and related defendants, and it’s a certainty that UniCredit’s lawyers at Skadden, Arps, Slate, Meagher & Flom will ask the judge to apply his ruling on Picard’s standing and bounce that suit as well.

Meanwhile, Judge Colleen McMahon, who is Judge Rakoff’s neighbor on the 14th floor of the Manhattan federal courthouse, is poised to rule on Picard’s standing in his common-law suits against UBS and JPMorgan Chase. McMahon is certainly an independent-minded judge so it would be a mistake to assume she’ll simply follow Rakoff’s lead. But Rakoff knew full well how intensely his ruling on Picard’s standing would be scrutinized, and nevertheless showed no equivocation in his opinion. It’s hard to imagine Judge McMahon reaching a contrary conclusion.